Introduction

What is Denoisr?

Denoisr is an AI-powered audio cleanup product built for spoken-word recordings. Upload an audio or video file, and Denoisr removes background noise — HVAC hum, fan noise, electrical buzz, room ambience — softens echo, and lifts the voice forward, while keeping the natural character of the speaker. From the same upload you can also generate a full transcript, summary, show notes, chapters, key takeaways, and ready-to-post social snippets.

No plugins, no manual noise profiles, no DAW. Upload, choose how you want it processed, and download. Sign up to claim 5 free credits (about 5 minutes of cleanup) — no credit card required.

What you can do with Denoisr

Denoisr packs three connected tools into one workflow:

• AI Audio Cleaner — Remove background noise, reduce echo, and enhance voice clarity on audio files (MP3, WAV, M4A, FLAC, OGG).
• Video Audio Cleaner — Clean the audio track of any video without re-encoding the picture. Original frames, codec, and resolution stay untouched; only the audio is replaced.
• Transcript & Show Notes Generator — Turn any recording into a speaker-separated, time-coded transcript plus a summary, podcast-ready show notes, chapter markers, key takeaways, and social posts — six deliverables in a single run.

A file you cleaned once can be sent to the Transcript Generator with one click — you don't need to upload it again.

How it works

Denoisr uses a neural network trained on large datasets of speech and background noise. When you upload a file, the model separates voice from non-voice sounds and produces a cleaned version. The processed audio is then re-muxed back into your original video if you used the Video Audio Cleaner.

Most short files (under 10 minutes) finish in a minute or two. Longer recordings and busy queue times take a little more, but a live progress indicator shows you where it is the whole way through. You can also close the tab — processing keeps running on the server, and your result is waiting for you in Cleanup History.

Three cleanup modes

Every audio or video run lets you pick one or more cleanup modes. Combining modes does not cost extra credits — it just takes a little longer.

• Clean Noise — Lightweight denoise. Strips background noise, fans, and room tone while preserving the original voice character. Good for already-decent recordings that only need the floor cleaned up.
• Enhance Voice (recommended for most recordings) — Full studio-grade pass: noise removal, loudness balancing, and vocal polish. The default choice when you want the voice to sit forward and consistent.
• Podcast Ready — Everything in Enhance Voice, plus automatic removal of filler words ("um", "uh"), long silences, mouth sounds, breaths, and stutters. End-to-end episode cleanup in one pass. Runs deep speech recognition and can take several minutes.

Podcast Ready always includes Enhance Voice — you can't run filler-word removal without the cleanup pass that makes it sound natural.

Highlights worth knowing

• Hear the difference before you upgrade — Every processed file plays your original and the cleaned version side by side. Listen, pick the variant that fits your project, then download.
• Process full-length episodes in one upload — Up to 90 minutes on Creator, 4 hours on Pro. Whole interviews, complete courses, full webinars — no splitting into chunks and stitching audio back together later.
• Choose your output format per file — Same cleaned source, three output formats. MP3 for streaming and sharing, WAV when you'll re-edit in a DAW, FLAC for lossless archival — picked at download time.
• Keep music; clean the voice — Toggle Keep background music on for vlogs or recordings with intro/outro tracks. Denoisr leaves the music intact and only cleans the spoken segments instead of treating music as noise.
• Parallel batch processing — Drop multiple files in one batch; every file is processed fully in parallel. Batch size depends on plan: Free 1 · Starter 2 · Creator 5 · Pro 20.
• Global edge upload — Video and large media upload through Denoisr's global edge network, so even 1 GB files move quickly from anywhere in the world.
• Cleanup History — Every run is saved with file name, submission time, status, credits used, and the output. Re-download anytime within your plan's retention window without re-processing.
• Privacy by default — Your source files are deleted from our storage within 24 hours after processing. We never use your audio to train our models or for any other purpose beyond producing your result.

Supported use cases

Denoisr is most effective on steady-state (stationary) noise — background sounds that are consistent and don't change much over time:

• HVAC and air conditioning hum
• Computer and laptop fan noise
• Electrical interference (50/60 Hz hum)
• Room ambience and light reverb
• Low-level traffic noise

Common content formats:

• Podcasts — Interview recordings, solo episodes, remote guest tracks
• Online courses — HVAC and fan noise from screencasts and lesson recordings
• YouTube voiceovers — Clean narration without treating your recording space
• Remote interviews — Mismatched mics and rooms pulled into consistent, listenable quality
• Field recordings — Recordings made in imperfect environments
• Meeting and webinar recordings — Clearer playback and more accurate transcripts

It can reduce but not always fully remove variable noise — sounds that appear suddenly and unpredictably (a dog barking mid-sentence, a door slamming). For those, manual editing is typically still needed.

Denoisr is built around spoken-word content. Full music production, detailed mixing, and mastering still belong in a dedicated audio workstation.

Transcript & show notes generator

Pointed at any audio or video file, the Transcript Generator produces all of the following in one pass:

• Full transcript — Speaker-separated and time-coded. Export as plain .txt or .srt subtitles ready for YouTube, Vimeo, Premiere, Final Cut, or Resolve.
• Summary — A clean short paragraph describing what the recording is about. Useful for episode descriptions, internal handoffs, or scanning.
• Show notes — Long-form, Markdown-formatted, ready to drop into Apple Podcasts, Spotify, Substack, or your website.
• Chapters — Auto-detected chapter markers with timestamps for YouTube chapters or podcast players that support segments.
• Key takeaways — Three to seven bullet points capturing the most important moments.
• Social posts — Ready-to-share snippets crafted for different platforms.

Speaker diarization is built in — turns are grouped by speaker with timestamps, so interviews, panels, and multi-person meetings are easy to follow. The transcription engine auto-detects the spoken language and supports English, Spanish, French, German, Portuguese, Russian, Chinese, Japanese, and many more. Summaries and show notes come back in the same language as the recording.

If you already cleaned a file with Denoisr, the result page offers a one-click button to send it straight to the Transcript Generator — no re-upload, the cleaned audio is reused automatically and you only pay the transcription credits.

Credits and pricing

Denoisr uses a simple credit system: 1 credit = 1 minute of audio or video cleanup. Transcription costs 1.5 credits per minute. Duration is always rounded up to the next whole minute — a 61-second file costs 2 credits for cleanup or 3 credits for transcription. Each file has a 1-credit-per-file minimum, so five separate 22-second clips cost 5 credits total (not 2). If you want to save credits on lots of tiny clips, merge them into one file before uploading.

Subscription plans

Annual plans grant the full yearly balance upfront (monthly allowance × 12), valid 365 days from purchase. Save about 17% versus monthly billing. Built for creators with bursty workflows — use credits whenever you need them, no monthly reset.

Pay-as-you-go credit packs

When you don't need a subscription, buy a one-time pack instead. Credits are valid 365 days from purchase, no auto-renew.

• Starter Pack — 80 credits for $9 (about 80 minutes of cleanup)
• Creator Pack — 300 credits for $19 (~ $0.063 / credit, 30% cheaper than Starter Pack)
• Pro Pack — 800 credits for $39 (~ $0.049 / credit, 45% cheaper than Starter Pack)

File-size, per-file duration, and batch limits on a pack match the equivalent monthly tier (Starter Pack → Starter limits, Creator Pack → Creator limits, Pro Pack → Pro limits).

Feature availability by plan

• Audio cleanup — Available on every plan, including Free.
• Video cleanup — Starter and above.
• Transcript generation — Creator and above.
• Filler-word & silence removal (Podcast Ready) — All paid plans.
• Long-form files (over 90 minutes) — Pro only (up to 4 hours per file).

If you try to use a feature that's locked on your current plan, Denoisr shows an upgrade card before charging credits — nothing starts processing until you confirm.

File requirements

• Audio formats — MP3, WAV, M4A, FLAC, OGG · up to 500 MB per file
• Video formats — MP4, MOV, M4V, WebM, MKV · up to 1 GB per file
• Output formats (cleanup) — MP3, WAV, or FLAC, chosen per download
• Subtitle / transcript outputs — .txt, .srt, Markdown (.md)
• Per-file duration cap — Free 5 min · Starter 30 min · Creator 90 min · Pro 4 hours
• Files per batch — Free 1 · Starter 2 · Creator 5 · Pro 20 (every batch runs fully in parallel)

For best results, upload files with a consistent background noise level throughout. Recordings with extremely variable noise environments (e.g., switching between indoors and outdoors mid-recording) can produce uneven results.

Cleanup History

Every run — audio cleanup, video cleanup, or transcript — is saved to your Cleanup History with its file name, submission time, current stage, credits used, and download links. Open any task to:

• Re-listen to the before and after, or re-watch the cleaned video
• Re-download in MP3, WAV, or FLAC (cleanup runs) or .txt/.srt/.md (transcript runs)
• See the credit cost for the run
• Retry failed runs

Output files stay available for download according to your plan's retention window (Free 7 days · Starter/Creator 30 days · Pro 90 days). After that the download links stop working, but the task entry itself remains visible.

Account & billing

• Sign-in — Email + password, or Google / GitHub sign-in. Email verification is required before processing.
• Profile — Update your name and account details from Account.
• Plans & Billing — Switch plans, view your renewal date, see your current credit balance, and cancel anytime from Plans & Billing. Cancellation keeps you on the plan through the end of the current billing period.
• Credits — See current balance, how it was built (subscription grant, top-up pack, or signup credits), and when each batch expires.
• Payments & Invoices — Every charge is itemized for download; subscriptions are managed through Creem.
• Security — Change password and review session activity.

Refunds & billing questions — email support@imagetostl.pro. We aim to respond within 3 business days. Approved refunds go back to your original payment method through Creem. For annual subscriptions, refunds are prorated against the portion of credits already consumed.

Privacy & data handling

• Source files are uploaded to private storage, used only to produce your requested result, and deleted within 24 hours.
• Cleaned audio, cleaned video, and transcript text are kept on your account during your plan's retention window so you can come back and re-download. You can delete any run from Cleanup History at any time.
• We never use your audio to train our models or for any other purpose beyond fulfilling your request.

Getting started

1. Sign up — Email + password, or Google / GitHub. You receive 5 free credits (5 minutes) on registration.
2. Upload a file — Drag and drop on the homepage (audio), the Video Audio Cleaner (video), or the Transcript Generator (audio or video).
3. Review the cost — Denoisr reads the duration in your browser and shows the exact credit cost before you click Start.
4. Pick your modes & options — Choose one or more cleanup modes; toggle Keep background music if needed; pick output format.
5. Wait for processing — Watch live progress, or close the tab and come back to Cleanup History for the result.
6. Compare and download — Listen to before/after, pick the output variant that fits, download in your preferred format.

Tips for the best results

• Record in a quiet environment first. Noise removal works best when the background noise is low to begin with. Every improvement at the recording stage makes the output cleaner.
• Process before other edits. Run cleanup on your raw recording before applying compression, EQ, or normalization. Processing a file that's already been heavily compressed can raise artifacts.
• Use headphones when evaluating. Laptop speakers mask noise at low levels. Headphones reveal the actual quality of the output and let you catch artifacts you'd miss otherwise.
• Don't over-process. If the output sounds unnatural or "watery", try Clean Noise instead of Enhance Voice. The goal is to reduce background noise to an acceptable level, not to eliminate every trace at the cost of voice quality.
• Keep music in vlogs and intros. If your recording includes intentional music, turn on Keep background music — otherwise the cleanup model may treat the music itself as noise.
• Merge tiny clips. Each file is billed with a 1-credit minimum. If you have many 15-second voice notes, concatenate them into one file before uploading.

Troubleshooting

• "Insufficient credits" — Your remaining balance is shown next to the credit cost. Buy a pack, switch plans, or pick a shorter file.
• "This file is longer than your plan allows" — Your current plan has a per-file duration cap. Upgrade to a plan with a longer limit, or split the file before uploading.
• "Too many tasks running" — You hit your plan's concurrency limit. Wait for one of your in-flight runs to finish, or upgrade.
• "Connection lost while checking progress" — Cleanup keeps running on the server. Refresh the page later, or check Cleanup History for the result.
• Video shows "We could not read this video" — The container or codec is unusual. Re-mux to MP4 (a lossless container swap is fine — no re-encoding needed) and try again.
• Output sounds watery or over-processed — Switch from Enhance Voice to Clean Noise, or turn off Podcast Ready if you don't actually need filler removal.
• Download link expired — Output files are kept according to your plan's retention window. Re-run the file to get a fresh download.

Support

For product questions, billing help, or refund requests, email support@imagetostl.pro. We typically reply within 3 business days.